Black start method after blackout in low inertia power system

Abstract

The present invention discloses a black start method after a blackout accident of a low-inertia power system, comprising: step 1: establishing a multi-objective optimization model considering the access principle of thermal power units and VSC-HVDC, including an objective function based on maximizing the output power of the unit and minimizing the system recovery time, wherein the constraints of the multi-objective optimization model include startup power constraints, startup time constraints, power flow and node voltage constraints, and VSC-HVDC operation constraints; step 2: solving the multi-objective optimization model using the Dijkstra algorithm and the NSGA-II algorithm; step 3: sorting the optimal solution set obtained in step 2, and using the multi-attribute decision-making method to screen out the optimal black start scheme, and using it as the black start scheme after a blackout accident of a low-inertia power system. The present invention solves the problem in the prior art that it is difficult to accurately describe the black start situation after a blackout accident of a low-inertia power system.

Description

Technical Field

[0001] The invention belongs to the field of power system security defense and system recovery, and in particular relates to a black start method after a large power outage accident in a low-inertia power system. Background Art

[0002] With the increasing demand for electricity and the increasing attention paid to environmental protection, and in order to better respond to the country's carbon neutrality and carbon peak policies, the proportion of new energy such as wind and light in the power grid will continue to increase for a long time in the future. At the same time, in order to adjust the problem of uneven distribution of "source-load", my country has vigorously built high-voltage direct current transmission (HVDC) projects to promote the consumption of new energy through large-scale and long-distance power transmission. In addition to the access of renewable energy and high-voltage direct current grids, flexible alternating current transmission systems (FACTS) and AC variable frequency drive loads are also developing rapidly. The power electronics of power systems are becoming increasingly intensified, and more and more power grids are gradually transforming into low-inertia power systems. Equipment manufacturing and system operation are facing new challenges and opportunities.

[0003] After a blackout in a low-inertia power system, the recovery process takes a long time due to the large number of units in the system and poor coordination between units. Therefore, it is of great significance to study the coordinated restoration of power supply of the power system by different types of units for the rapid recovery of low-inertia power systems.

[0004] When formulating a black start plan after a major power outage in the power system, the operating characteristics of the units need to be considered. The characteristics of different units are not the same. The low-inertia power system contains multiple units such as thermal power units, wind power units and high-voltage direct current transmission. The existing black start plan only considers the constraints of a single type of unit when modeling, which is difficult to adapt to the situation where multiple types of units in the low-inertia power system work together to restore system power supply. Summary of the invention

[0005] The purpose of the present invention is to provide a black start method after a blackout accident in a low-inertia power system, which solves the problem that when modeling in the prior art, only a single type of unit constraint is considered, and it is difficult to accurately describe the black start situation after a blackout accident in a low-inertia power system.

[0006] The technical solution adopted by the present invention is:

[0007] The black start method after a blackout accident of a low-inertia power system includes the following steps:

[0008] Step 1: Establish a multi-objective optimization model considering the access principles of thermal power units and VSC-HVDC; the multi-objective optimization model includes an objective function based on maximizing the unit output power and minimizing the system recovery time, and the constraints of the multi-objective optimization model include startup power constraints, startup time constraints, power flow and node voltage constraints, and VSC-HVDC operation constraints;

[0009] Step 2: Use Dijkstra algorithm and NSGA-II algorithm to solve the multi-objective optimization model obtained in step 1 to obtain the Pareto optimal solution set;

[0010] Step 3: Sort the Pareto optimal solution set and use the multi-attribute decision-making method to select the optimal black start solution as the black start solution after a blackout accident in the low-inertia power system;

[0011] The objective function based on maximizing the unit output power and minimizing the system recovery time is:

[0012] (1)

[0013] Where: f 1 is the negative value of the total power generation of the started units within the optimization time; N G is the total number of power supplies that need to be restored; T The time for all power supplies to start; For power supply i exist t The active power generated at any moment; c i Indicates the unit i Whether to invest, 1 for yes, 0 for no; f 2 is the sum of the recovery time of the started units; For the previous unit to the unit i The time required for line restoration; For the crew i The time required from starting to closing the synchronous circuit breaker to start ramping up and transmitting power outward; For the crew i The time required from the start of the climb to maximum handling;

[0014] Unit output function for:

[0015] (2)

[0016] In the formula, For the crew i The start time of The time required for unit i to start from startup to synchronous closing and start climbing and transmitting power outward; For the crew i The time required from the start of the climb to maximum handling; For the crew i Maximum climbing rate; For the crew i Rated output;

[0017] The starting power constraint is:

[0018] (3)

[0019] In the formula, To optimize the starting power of the unit within the time; c i Indicates the unit i Whether to invest, 1 for yes, 0 for no; For grid-connected units i Active power generated during the optimized time; For the crew i Required starting power;

[0020] The startup time constraints are:

[0021] (4)

[0022] In the formula, For the crew i The start time of For the crew i Maximum critical hot start time; For cold start i Minimum critical cold start time;

[0023] The power flow and node voltage constraints are:

[0024] (5)

[0025] In the formula, , For each unit i Active power The upper and lower limits of , For each unit i Reactive power The upper and lower limits of n 0 is the total number of generators in the restored system; n L is the total number of lines in the restored system; , They are the system stable runtime nodes i Voltage Upper and lower limits; nb is the number of nodes in the restored system; For branch i Active power flowing upstream;

[0026] The VSC-HVDC operation constraints are:

[0027] (6)

[0028] Where: P s , Q s are the active power and reactive power injected into the AC system respectively; X is the equivalent commutating reactance; μ is the DC voltage utilization rate; M is the modulation ratio; U s is the AC bus fundamental voltage; U dc is the DC voltage rating; I dc,max is the maximum DC link current.

[0029] The present invention is also characterized in that:

[0030] In step 2, the Dijkstra algorithm is used to calculate the line restoration time in the multi-objective optimization model. Solve; use NSGA-II algorithm to solve the multi-objective optimization model, and finally obtain the Pareto optimal solution set.

[0031] Step 3 is as follows: first, sort the Pareto solution set according to the congestion degree, then remove the feasible solutions whose recovery paths do not contain wind turbine nodes, then sort the remaining solutions according to the unrecovered load in the system, remove the bottom 5 solutions, and finally select the solution that prioritizes the recovery of wind turbine nodes.

[0032] The beneficial effects of the present invention are:

[0033] The present invention first establishes a multi-objective optimization model taking into account the constraints according to the access principles of different units, then uses the Dijkstra algorithm and the NSGA-II algorithm to solve the model, and finally makes a multi-attribute decision in the Pareto optimal solution to obtain a black start scheme for thermal power units, wind power units and VSC-HVDC to participate in power grid restoration. The present invention models and solves by considering the access principles of three types of units, namely thermal power units, wind power units and VSC-HVDC. Compared with the traditional black start scheme containing only a single type of unit, it can more effectively reflect the black start situation after a low inertia power system accident, and provides a feasible black start scheme for low inertia power system after a major power outage accident. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a flow chart of a black start method after a blackout accident of a low-inertia power system according to the present invention;

[0035] Figure 2 It is a flow chart of solving the multi-objective optimization model of the present invention;

[0036] Figure 3 It is the equivalent IEEE39-node low-inertia power system diagram in Embodiment 1 of the present invention;

[0037] Figure 4 It is a Pareto optimal solution distribution diagram of Example 1 of the present invention. DETAILED DESCRIPTION

[0038] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0039] The black start method after a blackout accident of a low-inertia power system of the present invention is as follows: Figure 1 and Figure 2 The low-inertia power system under consideration includes three types of units: thermal power units, wind power units and VSC-HVDC. First, it is necessary to clarify the access principles of different units, and establish a multi-objective optimization model considering its constraints according to the access principles of different units, solve it, and finally obtain the best solution, including the following steps:

[0040] Step 1: Establish a multi-objective optimization model considering the access principles of thermal power units and VSC-HVDC; the multi-objective optimization model includes an objective function based on maximizing the unit output power and minimizing the system recovery time, and the constraints of the multi-objective optimization model include startup power constraints, startup time constraints, power flow and node voltage constraints, and VSC-HVDC operation constraints;

[0041] Step 2: Use Dijkstra algorithm and NSGA-II algorithm to solve the multi-objective optimization model obtained in step 1.

[0042] Step 3: Sort the Pareto optimal solution set obtained in step 2, and use the multi-attribute decision-making method to screen out the optimal black start plan, and use it as the black start plan after a major blackout in the low-inertia power system.

[0043] In step 1, when establishing a multi-objective optimization model, it is necessary to consider the access principles of the two types of units, namely, thermal power units and VSC-HVDC. According to the access principles of the above two types of units, and taking into account the operating characteristics of different units and power grids, a multi-objective optimization model is established based on the objective function of maximizing unit output power and minimizing system recovery time:

[0044] (1)

[0045] Where: f 1 is the negative value of the total power generation of the started units within the optimization time; N G is the total number of power supplies that need to be restored; T The time for all power supplies to start; For power supply i exist t The active power generated at any moment; c i Indicates the unit i Whether to invest, 1 for yes, 0 for no. f 2 is the sum of the recovery time of the started units; For the previous unit to the unit i The time required for line restoration; For the crew i The time required from starting to closing the synchronous circuit breaker to start ramping up and transmitting power outward; For the crew i The time required from the start of the climb to maximum handling;

[0046] Unit output function for:

[0047] (2)

[0048] In the formula, For the crew i The start time of The time required for unit i to start from startup to synchronous closing and start climbing and transmitting power outward; For the crew i The time required from the start of the climb to maximum handling; For the crew i Maximum climbing rate; For the crew i Rated output.

[0049] The black start scheme of the present invention needs to meet the following constraints: unit startup constraints, AC system operation constraints and VSC-HVDC steady-state operation constraints.

[0050] (1) Starting power constraint

[0051] (3)

[0052] In the formula, To optimize the starting power of the unit within the time; c i Indicates the unit i Whether to invest, 1 for yes, 0 for no; For grid-connected units i Active power generated during the optimized time; For the crew i Required starting power.

[0053] (2) Startup time constraints

[0054] (4)

[0055] In the formula, For the crew i The start time of For the crew i Maximum critical hot start time; For cold start i The minimum critical cold start time.

[0056] (3) Power flow and node voltage constraints

[0057] (5)

[0058] In the formula, , For each unit i Active power The upper and lower limits of , For each unit i Reactive power The upper and lower limits of n 0 is the total number of generators in the restored system; n L is the total number of lines in the restored system; , They are the system stable runtime nodes i Voltage Upper and lower limits; n b is the number of nodes in the restored system; For branch i Active power flowing upstream.

[0059] (4) VSC-HVDC operation constraints

[0060] (6)

[0061] Where: P s , Q s are the active power and reactive power injected into the AC system respectively; X is the equivalent commutating reactance; μ is the DC voltage utilization rate;M is the modulation ratio; U s is the AC bus fundamental voltage; U dc is the DC voltage rating; I dc,max is the maximum DC link current.

[0062] In step 2, Dijkstra's algorithm is used to calculate the line recovery time in step 1. The Dijkstra algorithm is a typical single-source shortest path algorithm, which is used to calculate the shortest path from one node to all other nodes. The main feature is that it expands outward from the starting point to the end point. Line recovery time For the crew j To the unit i The sum of the recovery time of the lines passed is solved by Dijkstra algorithm to minimize it.

[0063] unit j To the unit i Required line restoration time for:

[0064] (7)

[0065] In the formula, Indicates the previous unit j To the unit i The shortest recovery time of the line weight is the recovery time of the line passed, where and .

[0066] In step 2, the non-dominated sorting genetic algorithm (NSGA-II) is used to solve the multi-objective optimization model in step 1. The NSGA-II algorithm uses non-dominated solutions for quick sorting, and uses elite strategy and congestion to select the Pareto optimal solution set that makes each objective function as large (or small) as possible. It has good robustness and speed, and can overcome the shortcomings of traditional multi-objective optimization algorithms. It is suitable for quickly solving multi-objective problems. The multi-objective optimization model established in step 1 contains 2 objectives and 4 constraints, and the NSGA-II algorithm is very suitable for solving it.

[0067] Step 3 performs multi-attribute decision making on the top 10 feasible solutions of the Pareto solution set obtained in step 2. When selecting the optimal solution, it is necessary to consider three factors: congestion, wind turbine access principle, and unrestored load in the system. After screening all feasible solutions, the optimal solution is obtained, which is used as the black start solution after a blackout accident in the low-inertia power system. Specifically, the Pareto solution set is sorted according to congestion, and then the solution whose recovery path does not contain wind turbine nodes in the feasible solution is removed. Then the remaining solutions are sorted according to the unrestored load in the system, and the bottom 5 solutions are removed. Finally, the solution that prioritizes the recovery of wind turbine nodes is selected.

[0068] The principle of the present invention is that the method disclosed in the present invention follows the following three access principles:

[0069] (1) Wind turbine access principles

[0070] ① The system needs to be restored to a certain capacity scale when the wind turbine is connected. In the early stage of black start, the system grid is small, and the only unit that can participate in frequency regulation is the black start power supply. In order to ensure the stability of the system frequency after wind power is connected, the present invention believes that the black start power supply must successfully start at least one started unit to form a small black start system with two or more units running in parallel before wind power can be considered to be connected.

[0071] ② When the system is restored to the wind power grid connection point, wind power cannot be self-started and must be powered by the power grid to start. Therefore, wind power access to the grid must wait until the system grid is restored to the wind power grid connection point.

[0072] (2) VSC-HVDC access principles

[0073] ① VSC-HVDC needs to successfully start the inverter-side converter station during the black start phase, thereby driving the passive grid to start and stabilize the AC system voltage. Therefore, the inverter-side converter adopts a fixed AC bus voltage control strategy.

[0074] ② The VSC-HVDC rectifier side adopts a constant DC voltage and constant reactive power control strategy to facilitate the stable operation of the flexible DC system in the early stage of black start.

[0075] (3) Principles for selecting the units to be started

[0076] ① After the started power plant (unit) is successfully started, it will serve as the main power source to supply power to other power sources and loads in the system and undertake subsequent system recovery tasks. Therefore, large-capacity units in large thermal power plants that are close to important loads should be given priority.

[0077] ② When a main unit in the power grid resumes its power generation capacity, the units to be started need to be optimized and selected based on the actual situation of system recovery during the subsequent unit recovery process.

[0078] The method of the present invention can more effectively reflect the black start situation after the low-inertia power system accident, and provides a feasible black start solution for the low-inertia power system after the blackout accident.

[0079] Example 1

[0080] This embodiment verifies the effectiveness of the method proposed in the present invention, and uses the improved IEEE 39-node system as an example for verification. Assume that the AC busbar on the VSC-HVDC inverter side is connected to the 30 nodes and connected to the external network through the DC line. The modified New England 10-machine 39-node system is, for example Figure 1 As shown, the rated transmission power of the DC line P d =100 MW. There are 9 units started in the receiving grid. The starting parameters of each unit are shown in Table 1, and the recovery time of each line is shown in Table 2.

[0081] Table 1 Parameters of the started power supply

[0082]

[0083] Table 2 Recovery time of lines in the system

[0084]

[0085] According to the method of the present invention, step 1 and step 2 are executed to model and solve. The line recovery time is used as the edge weight, and the Dijkstra algorithm is used to search for paths to solve the line recovery time. Maximizing the unit output power and minimizing the system recovery time are used as multi-objective functions, and the NSGA-II algorithm is used to solve the unit recovery model. The crossover probability of the algorithm is 0.9, the mutation rate is 0.05, the population size is 200, and the maximum number of iterations is 50. The distribution of the Pareto optimal solution obtained after solving the multi-objective optimization model is as follows: Figure 3 As shown. Figure 3 It can be seen that the algorithm in this paper has a good convergence effect.

[0086] Execute step 3: By calculating the examples and sorting the results according to the congestion degree, the top 10 feasible solutions are obtained as shown in Table 3.

[0087] Table 3 The top 10 feasible solutions in the Pareto solution set

[0088]

[0089] Multi-attribute decision-making is performed on the top 10 feasible solutions in the Pareto solution set, and different paths are analyzed and calculated. The different attributes of feasible solutions can be obtained as shown in Table 4.

[0090] Table 4 Different attributes of feasible solutions

[0091]

[0092] Firstly, according to the access principle of wind turbines participating in black start, the recovery paths in the feasible solutions are analyzed. When selecting the optimal solution, it is necessary to consider whether the nodes containing wind farms are restored. Among them, the recovery path of Scheme 2 does not contain 4 nodes, so Scheme 2 is excluded; then, according to the load recovery principle, the recovery paths in the feasible solutions are analyzed. When selecting the optimal solution, it is necessary to ensure that as many loads as possible are restored. Among them, there are many unrestored loads in Schemes 1, 5, 6, 9, and 10, so Schemes 1, 5, 6, 9, and 10 are also excluded; finally, according to the principle of giving priority to restoring nodes containing wind turbines, the remaining solutions are screened. Among them, the recovery time of nodes 4 and 16 in Scheme 7 is shorter than that of Schemes 3, 4, and 8. Therefore, after multi-attribute decision-making on the top 10 feasible solutions in the Pareto solution set, it can be concluded that the optimal black start solution is Scheme 7.

Claims

1. A black start method after a blackout in a low-inertia power system, characterized in that: The following steps are involved: Step 1: Establish a multi-objective optimization model considering the access principle of thermal power units and VSC-HVDC; the multi-objective optimization model includes an objective function based on maximizing the unit output power and minimizing the system recovery time, and the constraints of the multi-objective optimization model include startup power constraints, startup time constraints, power flow and node voltage constraints, and VSC-HVDC operation constraints; Step 2: Use Dijkstra algorithm and NSGA-II algorithm to solve the multi-objective optimization model obtained in step 1 to obtain the Pareto optimal solution set; Step 3: Sort the Pareto optimal solution set and use the multi-attribute decision-making method to select the optimal black start solution as the black start solution after a blackout accident in the low-inertia power system; The objective function based on maximizing the unit output power and minimizing the system recovery time is: （1） Where: f 1 is the negative value of the total power generation of the started units within the optimization time; N G is the total number of power supplies that need to be restored; T The time for all power supplies to start; For power supply i exist t The active power generated at any moment; c i Indicates the unit i Whether to invest, 1 for yes, 0 for no; f 2 is the sum of the recovery time of the started units; For the previous unit to the unit i The time required for line restoration; For the crew i The time required from starting to closing the synchronous circuit breaker to start ramping up and transmitting power outward; For the crew i The time required from the start of the climb to maximum handling; Unit output function for: （2） In the formula, For the crew i The start time of The time required for unit i to start from startup to synchronous closing and start climbing and transmitting power outward; For the crew i The time required from the start of the climb to maximum handling; For the crew i Maximum climbing rate; For the crew i Rated output; The starting power constraint is: （3） In the formula, To optimize the starting power of the unit within the time; c i Indicates the unit i Whether to invest, 1 for yes, 0 for no; For grid-connected units i Active power generated during the optimized time; For the crew i Required starting power; The startup time constraint is: （4） In the formula, For the crew i The start time of For the crew i Maximum critical hot start time; For cold start i Minimum critical cold start time; The power flow and node voltage constraints are: （5） In the formula, , For each unit i Active power The upper and lower limits of , For each unit i Reactive power The upper and lower limits of n 0 is the total number of generators in the restored system; n L is the total number of lines in the restored system; , They are respectively the system stable runtime nodes i Voltage Upper and lower limits; n b is the number of nodes in the restored system; For branch i Active power flowing upstream; The VSC-HVDC operation constraints are: （6） Where: P s , Q s are the active power and reactive power injected into the AC system respectively; X is the equivalent commutating reactance; μ is the DC voltage utilization rate; M is the modulation ratio; U s is the AC bus fundamental voltage; U dc is the DC voltage rating; I dc,max is the maximum DC link current.

2. The black start method after a blackout accident of a low-inertia power system according to claim 1, characterized in that: In step 2, the Dijkstra algorithm is used to calculate the line restoration time in the multi-objective optimization model. Solve; use NSGA-II algorithm to solve the multi-objective optimization model, and finally obtain the Pareto optimal solution set.

3. The black start method after a blackout accident of a low-inertia power system according to claim 1, characterized in that: The step 3 is specifically as follows: first, the Pareto solution set is sorted according to the congestion degree, and then the solution whose recovery path does not contain the wind turbine node in the feasible solution is removed, and then the remaining solutions are sorted according to the unrecovered load in the system, and the bottom 5 solutions are removed, and finally, the solution containing the wind turbine node is selected for priority recovery.

Images